This project will examine spatial cognition in foraging tasks with lexigram-trained chimpanzees at the Language Research Center, Georgia State University. Using protocol that has proven successful in past experiments, multiple food items will be hidden by the experimenter in a woodland visible but not accessible to the chimpanzees. An individual subject will watch, from an outdoor enclosure, as items are hidden. The chimpanzee will then later be able to, in a room with no view of the woodland, recruit a naive caretaker to recover the food items. Recruitment may involve a combination of lexigram-use, gesture, and vocalization. The interval between hiding and recruitment will vary between 15 minutes to imposed overnight delays. Subjects will then be given the opportunity to direct the uninformed person to the test items by whatever means necessary, including manual pointing, other gestures, and vocalization. Previous data has shown that chimpanzees will rank-order the recovery of hidden food items by type, quantity and distance from the outdoor enclosure. Several other ecologically-salient variables, taken from foraging theory and long thought to be integral to economic decision-making, will be investigated. In the first experiment, it is predicted that chimpanzees will rank order the recovery of food items by the handling time associated with them. Chimpanzees will also be trained to identify variable versus stable rewards with equal means using marked containers. It is expected that chimpanzees will exhibit preference based on variance (risk-sensitivity) and that these preferences will be extended to the rank-order recovery of items in the hidden foods experiment. In another experiment, large numbers of foods will be hidden in multiple aggregations (simulated "patches"). It is expected that chimpanzees will be able to remember aggregation size in their recovery of food items and exploit patches as predicted by an ecological model, the marginal value theorem. Lastly, the study will investigate the ability of chimpanzees to remember the "where" and "when," and multiple components of the "what" of an event, and will add significant data to the ongoing debate of whether any nonhuman animals possess an episodic memory system. This study would represent the first attempt to investigate many of these variables in the context of long-term memory. In addition, the experiments detailed are analogous to situations in which humans who lack speech must communicate their wants and desires, and provide directions for the study of memory in such individuals. The research also includes rare testing of recall memory in a nonhuman species, and is related to the concepts of prospective and episodic memory, all of which are particularly susceptible to decline in humans due to aging, injury, or disease.